Drug Induced Liver Injury
Robert J. Fontana, MD
University of Michigan Medical Center
DILI
• Idiosyncratic DILI
– – – – Epidemiology Clinical manifestations Causality assessment Prognosis
• Genetics in DILI
– Drug Induced Liver Injury Network
• Acetaminophen hepatotoxicity
– Epidemiology & prevention
Drug Induced Liver Injury
• DILI is the most common reason for regulatory actions concerning drugs
– Denial: Exanta – Withdrawal: Nefazadone – Warnings: Ketek
• Significant morbidity & mortality
– Leading cause of ALF in the US
• Acetaminophen 40% Idiosyncratic 13%
– Hepatocellular jaundice 10% mortality
• No reliable means to predict/ prevent
Hepatic Adverse Event Nomenclature
• Liver injury
– ALT > 3X ULN or – Alk phos > 2X ULN or – T Bili > 2X ULN + elevated ALT or alk – “Abnormality of liver tests” all others
• DILI categories
– Hepatocellular R > 5 – Cholestatic R < 2 – Mixed 2 < R < 5
R = (ALT/(ULN)) (Alk/ (ULN))
(HAEN Working Group 2005)
Liver Injury Classification
• Acute liver injury: < 6 mon +/- symptoms • Chronic liver injury: > 6 mon +/- symptoms • Severe liver injury
– Jaundice (Bili > 2 xULN) – INR > 1.5 xULN – Encephalopathy
• Fulminant liver injury
– Coagulopathy and encephalopathy within 4 weeks
(HAEN Working Group 2005)
DILI is uncommon
Seattle 1 ‟77-81 Design # at risk
Retro, HMO database 280,000
Mass 2 ‟92-93
Retro, HMO ICD-9 code 160,000 inpt + outpatient
Swiss 3 ‟96-00
Retro, 1 center 4,209 inpatient
France 4 ‟97-00
Prospect cohort 81,301 inhabitants
Inclusion
# DILI case Incidence
Hospitalize
12 .001%
LFT‟s
50 0.04%
LFT‟s
57 1.4%
All DILI
34 0.014%
(1 J Clin Pharm 1986; 26: 633) (2 Pharmacoepi & Drug Safety 1999; 8: 275) (3 Eur J Clin Pharm 2005; 61: 135) (4 Hepatology 2002; 36: 451)
DILI Population based study
81,300 in France ‟97-‟00 • 95 suspected DILI cases
– 34 probable DILI
• 25% antibiotics 23% psychotropic 13% hypolipid • 80% outpatients • 2 (7%) deaths
– 61 other causes/ inadequate data
• Incidence: 14 to 24 per 100,000
– 8,000 annual cases, 500 deaths – 16 X > than ADR surveillance
(Hepatology 2002; 36)
DILI Diagnosis
• DILI is a diagnosis of exclusion based on circumstantial evidence due to lack of confirmatory lab test, rechallenge, or “GOLD” standard
• DILI diagnosis is invariably retrospective
– Exclude other causes – Dechallenge requires follow-up – Requires a high index of suspicion
Clinicopathologic forms of DILI
• Acute hepatitis • Acute cholestasis • • • • • • • Chronic hepatitis Fatty liver/ NASH Granulomatous hepatitis Fibrosis/ cirrhosis Vanishing bile duct VOD, peliosis Benign & malignant neoplasia
DILI: A Diagnosis of exclusion
• Temporal relationship
– Latency: usually < 12 mon – Not dose related – ? Clinical risk factors
• Biochemical injury pattern
– “Signature” vs protean – Prior reports/ cases – Exclude other likely causes
• Improve with discontinuation
Acute Hepatocellular: Differential Dx
Ultrasound/ CT
(A, B, C, CMV, EBV HEV, HSV)
Viral
++
Autoimmune
(SPEP, ANA, SmAb)
NAFLD
Mass
(AFP, MRI)
Biliary
(ERCP)
Ischemia
(History, 2D-Echo) Observe/ biopsy
Metabolic
(Iron, TIBC, ferritin, ceruloplasmin, SPEP)
DILI ( <1%)
(Gordon J Clin Gastro 2005; 39: 64)
ADR: Causality Assessment
100%
● Definite ● Highly probable ● Probable
50%
● Possible
● Unlikely ● Excluded / other
0%
DILI: Causality Assessment
• Generic instruments
– WHO – Bayesian
• Liver specific
– Expert opinion – Roussel Uclaf Causality Assessment Method (RUCAM) „89 – Clinical Diagnostic Scale (CDS) „97
RUCAM
• • • • • • • Temporal relationship Course Risk factors Concomitant drug Non-drug causes Prior reports/ information Re-challenge (0 to 2) (-2 to 3) (0 to 2) (0 to -3) (-3 to 2) (0 to 2) (-2 to 3)
Score (- 8 to 14) Highly probable >8 Probable 6-8
Possible 3-5 Excluded ≤0 Unlikely 1-2
J Clin Epidemiol 1993;46:1323-1330
RUCAM limitations
• Ambiguous instructions
– Criteria for competing cause/drug not clear – Onset > 30 days after d/c (e.g. Augmentin)
• Derived from expert opinion rather than prospectively collected data set
– Limited risk factors – Overweighting of rechallenge
• Low inter-observer reproducibility
228 Spanish cases ‟94-‟00
Gold standard: Expert panel
RUCAM Exclude Unlike Poss Prob Definite
Exclude 21 4 1
CDS Unlike 2 3 8 30 5
Poss
Prob Def
1 43 40
16 53
1
K = 0.28
* 30 non-drug cases
Poor performance in ALF/ Death (6%)
CDS: 6 possible 7 unlikely/ excluded 6 probable 1 possible
(Hepatology 2001; 33: 123)
RUCAM: 6 definite
Prognosis in DILI with jaundice
836 Swedish cases (‟70-04) Total (784)
Med Age % Female % Died/ txp 58 58% 9.2%
HC (409)
55 58% 12.3% *
Chol (206)
69 59% 8.7%
Mix (169)
59 57% 2.4%
* P < 0.05 vs mixed and mixed + cholestatic
Age, AST, bilirubin predict death/txp
However, suspect drug also important
- 0% erythromycin 40% halothane
(Bjornson Hepatology 2005)
Spectrum of DILI
ALF (Death, Txp) 0.0001 - 0.01%
Symptomatic disease 0.01 - 1.0% Mild liver injury (ALT < 3X ULN) 0.1 - 10%
ALT monitoring and DILI
• INH causes ALT 1-3X ULN in 20%
– ? Adaptation with continued treatment
• INH hepatitis: ALT > 3X ULN + jaundice 1-2%
– 50% in first 2 months, non-specific symptoms – Most resolve with INH discontinuation
• Fulminant hepatitis < 0.1%
– 177 deaths due to INH 19921 – Leading cause of DILI-ALF leading to LT 2 – Age > 40 ? Female ? Alcohol/ liver disease
• Monitor serum ALT levels on INH?
– In who? Frequency ? Criteria to discontinue ? – Alternate regimen ? Rif + PZA contraindicated
(1 Am Rev Respir Dis 1992; 145: 494) (2 Liver Transplantation 2004; 10; 1018)
ALT monitoring with INH
• CDC recommends baseline and regular LFT monitoring in
– HIV +, pregnancy – Liver disease, selected elderly
• Monthly “clinical” assessment and labs if symptoms
– Individualize monitoring for active TB
(JAMA 1999; 281: 1014)
“Idiosyncracy”
Hippocrates, ~400 B.C. (idios) - one‟s own, self
(syn) – together (crasis) - a mixing, mixture
A person‟s own mixture of characteristics, factors, nature and nurture, uniquely
John Senior - FDA
DILI pathogenesis
Drug
Class Dose Duration
Environment
Diet, toxins and exposures (tobacco, alcohol, coffee, chemicals, pollutants, oxidants, probiotics)
Host
Age, gender, weight, genetic factors, immune factors, other diseases
Drug Induced Liver Injury Network
A cooperative Agreement funded by the Division of Digestive Diseases and Nutrition
National Institute of Diabetes and Digestive and Kidney Diseases
DILIN: Sphere of Influence
Indiana U
N. Chalasani
U Michigan
R. Fontana
U Conn
UCSF
T. Davern H. Bonkovsky
U N Carolina
P. Watkins
12.8 million lives
Prospective study
Multicenter, longitudinal study of Drugand CAM- induced liver injury
Retrospective study
Idiosyncratic Liver Injury Associated with Drugs (ILIAD)
Prospective study - AIMS
• #1 To identify bona fide cases of drug and CAM induced liver injury within 6 months of presentation so that clinical data and samples can be collected for future mechanistic and genetic studies • #2 To identify clinical, immunological, and environmental risk factors for drug and CAM liver injury by comparing cases to controls
Prospective Study
• • • • Liver injury within 6 months Medications and CAM Children and adults Pre-defined eligibility criteria for cases and controls • Distinct entry criteria for patients with underlying liver disease (HCV, HBV) • Minimum of 6 months follow-up
– Maximum of 24 months follow-up
• Detailed data collection
Inclusion criteria
• Age > 2 • DILI within 6 months of presentation • On 2 consecutive blood draws
– AST/ ALT > 5 X ULN or baseline – Alk phos > 2 X ULN or baseline – T bilirubin > 2.5 mg/dl
• Chronic HBV, HCV, HIV allowed
DILIN Causality Committee
5 site PI‟s, 1 DCC, 1 NIH
• 3 independent reviewers • Review clinical narrative, subset CRF
– Score causality (25% > 1 drug) – RUCAM – Data completeness checklist
• Conference call to finalize
DILIN Causality assessment
• • • • • • Likelihood > 95% 75 -95% 50 -75% 25 -50% < 25% Category Definite Likely Probable Possible Unlikely
DILIN - Prospective and Retrospective Monthly Cumulative Patient Enrollment
250 240 230 220 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0
Jun-04 Jul-04 May-04
Observed Curve for Retrospective Study Observed Curve for Prospective Study
Number of Patients Enrolled
Nov-04
Nov-05
Aug-04
Aug-05
Aug-06
Nov-06
Apr-05
Apr-06
Oct-04
Oct-05
Jun-05
Jun-06
Dec-04
Feb-05
Dec-05
Feb-06
Oct-06
Jul-05
Jul-06
Sep-04
Sep-05
May-05
Month of Study
May-06
Sep-06
Dec-06
Jan-05
Mar-05
Jan-06
Mar-06
Implicated drugs (N=141)
Single prescription drug Single CAM 75% 3.9%
Multiple drugs/ CAM Hepatocellular Cholestatic Mixed
21.1% 55.9% 20.3% 23.7%
143 Prospective cases
52 Antibacterial 14 Anticonvulsant 12 AntiTB (6) / fungal 9 Psychiatric/ ADHD 9 CAM products 8 Biological 8 NSAID 6 Hypolipidemic 6 Cardiovascular 4 Antineoplastic 4 General anesthesia 4 Antiviral 2 STUDY drugs 5 Others
New agent signals
• • • • • • Arava (leflunomide) (2) Enbrel (etanercept) (2) Stratera (atomoxetine) (2) Telethromycin (ketek) (3) Rebif (IFNB-1a) (2) Research agents (2)
Demographics (n=141)
Female Cau AA Other Mean age (range) BMI Hospitalized 60.3% 78.6% 14.3% 9.3% 48.7 (2-83) 26.8 (14-50) 63.4%
Drug
Stable metabolites, excretion
detoxification
bioactivation
reactive metabolite
immune mechanisms
Non-immune mechanisms
Hepatocyte damage
DILIN Scientific AIMS
• HYPOTHESES: Variation in host drug metabolizing, detoxification, or regeneration/ adaptation pathways may explain (in part) susceptibility to DILI • METHODS: Collect DNA and compare DILI cases vs controls
Human genome
• Human genome: 30,000 unique genes • 10 x 10 6 single nucleotide polymorphisms
– May influence susceptibility or outcome with disease and/ or drugs
• ~ 50% of interindividual variation
– Multiple SNP‟s likely required
• Whole genome scanning
– High throughput technology available – 500,000 SNP‟s across entire genome
• Studies require well “phenotyped” cases
1
2
3
4
5
1 1 2 2 3 3 4 4 5 5
DILI susceptibility
High risk: Homozygous for red/ dysfunctional SNP‟s on > 50% of genes Low risk: <50% single dysfunctional SNP‟s
DILIN Genotyping Initiative
• Collaboration with GlaxoSmithKline
– Affymetrix Gene chip for 500k SNP‟s – Analyze DILI cases vs 500 population controls matched by age, race, ethnicity – 50k non-synonomous SNP chip – 1,300 Absorption, Distribution, Metabolism and elimination genes – 5,000 markers in candidate genes
• Replicate genetic associations in other populations, investigate pathways, etc
Acetaminophen: Friend or foe ?
• Safe & effective
– > 1 billion tabs / yr
• Preferred to ASA in liver dz, children
– 200 OTC products & > 20 Rx drugs
• Hepatotoxicity
– Dose dependent (> 4 grams) – > 60,000 overdose/ yr
• Impulsive suicide gestures • 500 deaths/yr
Glucuronyl transferases sulfotransferases Acetaminophen stable metabolites, excretion
Glutathione transferases
CYP2E1, CYP3A4, CYP1A2
NAPQI
Hepatocyte damage
% Acetaminophen ALF in the US
To tal ALF cases:
85
94
99
123
133
128
140
Perc ent of ALF Case s
60% 50% 40% 30% 20% 10% 0% 28% 38% 44% 38% 47%
51% 51%
1998 1999 2000 2001 2002 2003 2004 YEAR
(W Lee ALFSG 2005)
ACM related ALF in the US
Intentional Non-intentional
(n=122) (n=131)
% Female % Pain ACM (g) / d Total ACM (g) % > 2 ACM formul. % Narcotic-ACM % Alcohol
% Survival
* P < 0.05
74% 0 29 29 5% 18% 59%
71%
73% 82% * 10 * 34 38% * 63%* 53%
72%
(Larson et al Hepatology 2005)
ACM-cysteine adducts
1.0
Pt 1, Admisssion APAP = 90 mg/dL
7000
1.0
Pt 4, Admission APAP = 0 mg/dL
1600 1400
APAP-CYS (nmol/mg protein)
6000 0.8 5000
APAP-CYS (nmol/mg protein)
0.8 1200
ALT (IU/L)
0.6
4000
0.6
800 0.4 600 400 0.2 200 0.0 1 2 3 4 5 6 7 0
0.4
3000
2000 0.2 1000
0.0 1 2 3 4 5 6 7
0
Hospital Day
Hospital Day
Intentional
ACM overdose
Non-intentional
ACM overdose
Biomarker for ACM hepatotoxicity
- Sensitive & specific for ACM liver damage - 19% of “indeterminate” ALF
(Davern US ALFSG 2005)
ALT (IU/L)
1000
Implications of ACM related ALF
• ALF due to ACM is increasing over time
– > 50% non-intentional “misadventures” – Late presentation/ low serum ACM levels
• High index of suspicion for early NAC
– ? Adduct assay ? Prognostic criteria
• Regulatory actions
– Dispensing, packaging, label of OTC – ? Composition of narcotic-ACM
Acetaminophen (4 g/d) x 14 days in healthy volunteers
ALT > 3X ULN 0% 38% 41% 44% 31% ALT > 5X ULN 0% 23% 19% 37% 25%
Placebo (39) ACM (26) ACM + oxycodone (27) ACM + hydromorph (27) ACM + morphine (26) 56% Hispanic Mean age= 33 ? Adducts
? Adaptation ? Generalizability
(Watkins JAMA 2006; 296: 87-93)
Acetaminophen advice
• ACM is a safe and effective analgesic • American Liver Foundation July 18 2006
– “Do not exceed 3 g/d for prolonged time” – “Liver dz patients should check with their doctor”
• To avoid inadvertent toxicity
– Read labels & monitor total dose – ? Change narcotic-ACM congeners
DILI in 2006
• DILI is uncommon and difficult to diagnose
– Annual incidence of ~ 1 in 10,000 person years
• < 1% of acute liver injury
– Diagnosis of exclusion
• However, DILI is important
– Most common reason for FDA actions – 10% mortality if hepatocellular jaundice • Prospective surveillance networks (DILIN) may improve understanding of host, environmental, and genetic risk factors of “well-phenotyped” cases
DILIN Initiatives
• AASLD STC on DILI (Sept 2005)
– Hepatology 2006; 43: 618-631
• FDA/PHARMA/ AASLD Steering committee
– Jan 24, 2006- RFA for industry collaboration
• National Library of Medicine (J Snyder)
– LIVERTOX database on the web – Annotated references re causality – ? Post DILIN cases
• Genetic polymorphism analysis
Acetaminophen toxicity in severe acute HAV/ HBV
• Sera from 72 HAV/ HBV ALFSG patients assayed for ACM-protein adducts
– Adduct + 5/49 (10%) HBV (0.4 nmol/l) – Adduct + 4/23 (17%) HAV – Adduct + 10/10 (100%) ACM OD (5.6 nmol/l)
• 8 of 9 patients reported some ACM use
– All < 4 grams/ day
• 67% of adduct + died vs 27% adduct (-)
(#S1002 Polson DDW 2006)
Indeterminate
Acetaminophen-CYS (umol/L)/mg protein
3.0
Known APAP
with adducts N=7
nmol APAP-CYS / mg protein
2.5
2.0
1.5
1.0
Other ALF APAP No tox
0.5
Indeter minate N=29
0.0 A B C D E
Patient Group
Davern TJ, et al. Gastroenterology 2006;130:687-94
Dispensing, packaging, label changes of ACM products in UK „98
12 mon 12 mon p Pre Post 9 9 81 x 10 45 x 10 0.0001 2186 309 49 25 185 1956 193 37 12 147 < 0.001 < 0.001 <0.001 < 0.001 0.01
(BMJ 2001; 322:1)
ACM sales OD cases LT referral LT List LT done Deaths
287 Japanese DILI cases ‟78-‟02
55% hepatocellular 24% mixed 22% cholestatic
200
Number of Cases
150
100
50
0 Unrelated Unlikely Possible Probable Highly Probable
RUCAM
RUCAM + DLST
•Latency > 15 or 30 d changed, other drugs omitted, • + DLST = +2 Eosinophils > 6% = + 1
(Hep Research 2003: 27: 191)
Prognosis in Hepatocellular DILI
• Fulminant DILI has poor prognosis 1
– Acetaminophen (39%): 70% survival – Idiosyncratic DILI (13%): 25% survival
• Hy‟s rule: Hepatocellular DILI leading to jaundice has ~ 10% mortality 2
(1 Annals Int Med 2002; 137: 947) (2 Zimmerman Hepatotoxicity 1975)
Prospective Study Design
Case
<6 DRUG A 0 DILI Onset BL Visit 6 mon F/u 12 & 24 mon F/u 6 6 mon
Control
DRUG A 0 BL Visit
DILI Causality Instrument
• • • • • Sensitive Specific- Low probability in non-drug cases Reproducible Content validity- weighting is evidence based Criterion validity- “Gold Standard” expert panel Discrimination- a semi-quantitative estimate Validated in independent groups Generalizability- Young vs old, mild vs severe, hepatocellular vs cholestatic, normal vs abnormal baseline LFT‟s Ease of use
• • •
•
Baseline features (n=141)
Hospitalized Extrahepatic manifestations Liver biopsy Liver transplant Steroids Illness duration < 1 week 1 -2 weeks 2-4 weeks > 4 weeks 63.4% 10.5% 46.7% 2.2% 17.8% 8.1% 2.2% 28.9% 60.7%
400 350
US ALFSG „98-‟05 n=838 adults
45%
100%
80% 300 250 200 150 100 20% 50 0
A H A k oc Sh Pr B O O g n ru D he op in m H on il s W ta ce at ep s iti B m oi ut C dud v er th er th at ep y nc na eg
% Spontaneous survival
60%
12%
40%
0%
e at in rm te de In
s iti
e un m H at ep
s se e m iru ro nd Sy
i ar hi
D
se ea is
A
Transplantation-free survival Transplantation Died before Transplantation Transplantation-free survival rate
s iti
ACM related ALF in the US
Serum ALT Serum ACM % King‟s Intentional (n=122) 5326 95 7% Non-intentional (n=131) 3319 * 42 20%
* P < 0.05
(Larson Hepatology 2005)
DILI Clinical features
France1 Sweden2 ‟97-00 ‟70-04 Japan3 ‟78-02 Spain4 ‟94-04
n
Age % Female
34
> 50 65%
784
58 58%
287
NA NA
461
53 49%
% Hepatocellular % Mix/ cholest % Hosp % Death or txp
52% 47% 12% 6%
52% 22 /26% NA 9.2%
55% NA NA
58% 53% 7%
23 /22% 22/20%
(1 Hepatology 2002; 36: 451) (2 Hepatology 2005; 42: 481) (3 Hepatology Research 2003; 27: 192) (4 Gastroenterology 2005; 129: 512)
Prognosis in DILI with jaundice
• 836 presumed DILI cases with jaundice from Sweden („70-04)
– 454 hepatocellular (54%) – 213 cholestatic (25%) – 169 mixed (21%)
• AIM: Verify if “Hy‟s rule” applies to consecutive cases of severe hepatocellular injury and identify predictors of death/ transplant
(Bjornsson Hepatology 2005)
Regulatory actions due to DILI
(1995-2005)
Withdrawals bromfenac troglitazone Second Line felbamate pemoline tolcapone trovafloxacin Warnings
acetaminophen leflunomide nefazodone nevirapine pyrazinamide/rifampin terbinafine valproic acid zifirlukast atomoxetine saquinavir/rifampin Interferon 1a infliximab (kava, lipokinetix)
http://www.fda.gov/medwatch/safety.htm